C++: fix trigraph parsing in macros.

Trigraphs must only be parsed before/during preprocessing. The preprocessor
will now replace trigraphs with their standard form, and re-lexing in
TranslationUnit will not try to parse any trigraph.

Also added a few missing trigraphs: ??=, ??', ??! and ??-.

Task-number: QTCREATORBUG-13253
Change-Id: I1723ed53b00090b878c22b83b7e963b647b65f72
Reviewed-by: Nikolai Kosjar <nikolai.kosjar@theqtcompany.com>

3 files changed, 47 insertions, 2 deletions

diff --git a/src/libs/3rdparty/cplusplus/Lexer.cpp b/src/libs/3rdparty/cplusplus/Lexer.cpp
index db7cb9e0f8..68f87f126c 100644
--- a/src/libs/3rdparty/cplusplus/Lexer.cpp
+++ b/src/libs/3rdparty/cplusplus/Lexer.cpp
@@ -336,20 +336,62 @@ void Lexer::scan_helper(Token *tok)
         break;
 
     case '?':
-        if (_yychar == '?') {
+        if (_yychar == '?' && f._ppMode) {
             yyinp();
             if (_yychar == '(') {
                 yyinp();
                 tok->f.kind = T_LBRACKET;
+                tok->f.trigraph = true;
             } else if (_yychar == ')') {
                 yyinp();
                 tok->f.kind = T_RBRACKET;
+                tok->f.trigraph = true;
             } else if (_yychar == '<') {
                 yyinp();
                 tok->f.kind = T_LBRACE;
+                tok->f.trigraph = true;
             } else if (_yychar == '>') {
                 yyinp();
                 tok->f.kind = T_RBRACE;
+                tok->f.trigraph = true;
+            } else if (_yychar == '=') {
+                yyinp();
+                tok->f.trigraph = true;
+                if (_yychar == '?' && *(_currentChar + 1) == '?' && *(_currentChar + 2) == '=') {
+                    yyinp();
+                    yyinp();
+                    yyinp();
+                    tok->f.kind = T_POUND_POUND;
+                } else {
+                    tok->f.kind = T_POUND;
+                }
+            } else if (_yychar == '\'') {
+                yyinp();
+                if (_yychar == '=') {
+                    yyinp();
+                    tok->f.kind = T_CARET_EQUAL;
+                } else {
+                    tok->f.kind = T_CARET;
+                }
+                tok->f.trigraph = true;
+            } else if (_yychar == '!') {
+                yyinp();
+                if (_yychar == '=') {
+                    yyinp();
+                    tok->f.kind = T_PIPE_EQUAL;
+                } else {
+                    tok->f.kind = T_PIPE;
+                }
+                tok->f.trigraph = true;
+            } else if (_yychar == '-') {
+                yyinp();
+                if (_yychar == '=') {
+                    yyinp();
+                    tok->f.kind = T_TILDE_EQUAL;
+                } else {
+                    tok->f.kind = T_TILDE;
+                }
+                tok->f.trigraph = true;
             }
         } else {
             tok->f.kind = T_QUESTION;
diff --git a/src/libs/3rdparty/cplusplus/Lexer.h b/src/libs/3rdparty/cplusplus/Lexer.h
index d47dcdf12a..8e862ea54c 100644
--- a/src/libs/3rdparty/cplusplus/Lexer.h
+++ b/src/libs/3rdparty/cplusplus/Lexer.h
@@ -125,6 +125,7 @@ private:
         unsigned _scanKeywords: 1;
         unsigned _scanAngleStringLiteralTokens: 1;
         unsigned _ppMode: 1;
+        unsigned _ignoreTrigraph : 1;
     };
 
     struct State {
diff --git a/src/libs/3rdparty/cplusplus/Token.h b/src/libs/3rdparty/cplusplus/Token.h
index 39a74031c9..afe732d493 100644
--- a/src/libs/3rdparty/cplusplus/Token.h
+++ b/src/libs/3rdparty/cplusplus/Token.h
@@ -371,8 +371,10 @@ public:
         // The token is C++11 user-defined literal such as:
         // 12_km, 0.5_Pa, 'c'_X, "abd"_L, u16"xyz"_M
         unsigned userDefinedLiteral : 1;
+        // Indicates the token is a trigraph
+        unsigned trigraph      : 1;
         // Unused...
-        unsigned pad           : 2;
+        unsigned pad           : 1;
         // The token length in bytes and UTF16 chars.
         unsigned bytes         : 16;
         unsigned utf16chars    : 16;